# Importing Data
CAMERON<- read_excel("C:/users/Biniam/Desktop/Documents/Academic/Thesis/Analysis Folder/Excel Files/Aluminium/Aluminium.xlsx",sheet = "Sheet1", range = "H1:H229")

# Checking the Imported Data
View(CAMERON)
# Creating Time Series Data
CAMERON_ts <- ts(CAMERON, start=c(1998,1), end=c(2016,12), frequency=12)
# Viewing and Checking the Created Time Series Data
CAMERON_ts
sum(is.na(CAMERON_ts))
library(forecast)
CAMERON_ts <- tsclean(CAMERON_ts)
CAMERON_ts

# Identification: Plotting the Time Series Data
plot(CAMERON_ts)

# Estimating the appropriate model
CAMERON_ts_model <- auto.arima(CAMERON_ts)
CAMERON_ts_model

# Forecasting
options(max.print=1000000)
CAMERON_ts_forecast <- forecast (CAMERON_ts_model, level=c(95), h=288)
plot(CAMERON_ts_forecast)
CAMERON_ts_forecast             

# Exporting
write.table(CAMERON_ts_forecast, file="/users/Biniam/Desktop/Documents/Academic/Thesis/Result Folder/TSA Results/Excel Files/From R/Aluminium/CAMERON_TSA.csv", sep=",")
